Randomized Feasibility Trial of Routine Versus Selective Transesophageal Echocardiography During Isolated Coronary Artery Bypass Grafting

Abstract

Background: Intraoperative transesophageal echocardiography (TEE) is used in approximately half of isolated coronary artery bypass grafting (CABG) procedures in the United States, yet its routine use remains unsupported by randomized evidence. Objectives: The objective of the study was to assess the feasibility and protocol adherence of routine vs selective intraoperative TEE strategies during isolated CABG surgery. Methods: We conducted a pragmatic, randomized feasibility trial at 2 tertiary-care hospitals within a single health system. Adults undergoing isolated CABG for whom both strategies were clinically acceptable were randomized 1:1 to TEE-by-default (routine use) or TEE-on-demand (selective use). In the selective group, TEE was performed as clinically indicated. Primary feasibility outcomes included enrollment, successful randomization, and protocol adherence. Exploratory outcomes included perioperative clinical events, patient-centered recovery measures, and TEE-related adverse events. Results: A total of 274 patients were screened and 72 were eligible. Of these, 46 (64%) consented, and 40 (56% of eligible; 87% of consented) were randomized following induction of anesthesia (20 per group). Protocol adherence was high (39/40; 98%) with 1 allocation error. In the TEE-on-demand arm, 3 of 20 patients (15%) underwent clinically triggered TEE. No deaths occurred within 90 days. Three serious adverse events were adjudicated as related to TEE exposure including 2 esophageal complications and 1 conversion to combined CABG and mitral valve replacement. Conclusions: In this pragmatic randomized feasibility trial, both TEE strategies were successfully implemented with high protocol adherence and preserved clinician and patient equipoise, supporting a multicenter randomized trial evaluating routine vs selective TEE during isolated CABG.

Bo Zhang

Assistant Professor of Biostatistics

My research interests include design of observational studies, instrumental variables, application of causal inference in medicine and applied statistics in general.